Uso de las Herramientas Informáticas y del Internet en la Enseñanza de ELE. 59

In this study, I attempt to address the overarching question “How is peer-peer collaborative dialogue carried out in a SCMC environment and how does it influence ESL learners’ acquisition of the target English idioms?” This question, based on the

literature review, stems from the need for more empirical evidence regarding SCMC- based collaborative dialogue and L2 vocabulary gains from ESL learners’ languaging about the lexical items targeted for learning. Another motivation is that, while current studies on collaborative dialogue are in general quantitative, quasi-experimental in nature and seem mostly concerned with the linguistic properties (for example, the frequency and the focuses) of LREs, qualitative inquiry into the functions of collaborative dialogue, or more specifically the exploitation of L2 for language socialization and language learning, would undoubtedly offer additional insights. Furthermore, even in the few existing studies that investigated collaborative dialogue using a qualitative approach, language socialization and language learning were often examined separately. Given that

sociocultural SLA views L2 development as situated participation in which learners draw on L2 for both language socialization and language learning (Pavlenko & Lantolf, 2000; Zuengler & Cole, 2005; Zuengler & Miller, 2006), the integration of these two functions would give a richer, more comprehensive picture of collaborative dialogue than would a focus on any one function alone. Finally, as mentioned before, lexical items emerging from LREs or selected as the learning targets were largely confined to simple, concrete nouns or humorous and playful uses of language, and the acquisition of these items was usually investigated in relation to learners’ gain scores. There have thus been calls for more research that looks at lexical gains on vocabulary of higher level of difficulty through collaborative dialogue and a microgenetic account of how learners internalize co- constructed lexical knowledge through moment-by-moment interaction. The current study, to a great extent, serves the aforementioned purposes by gathering fine-grained data relating to the deciphering of the figurative meaning of the idioms that are frequently

used in academic discourse through real-time discussions. In offering a glimpse into the participants’ creation of a social space in SCMC in which they participated in languaging about the meaning of the target idioms, the current study aims to reveal how ESL learners made use of scaffolded assistance and appropriated “mediational means, such as language, made available as they interact in socioculturally meaningful activities” (Zuengler & Miller, 2006, p. 39) for the achievement of L2 lexical knowledge growth. Specifically, the research questions addressed in this study are:

1. What patterns of SCMC-based dyadic interaction do the participants engage in in deciphering the meaning of the target English idioms within the idiom-focused-dialogue (IFD) episodes?

2. What communication strategies do the participants utilize to manage and maintain their collaborative interaction within IFD episodes?

3. In what ways do the participants provide scaffolded assistance to each other during IFD episodes to achieve mutual comprehension of the meaning of the target idioms? 4. What are the characteristics of online collaborative interaction that dyads with high and low scores exhibited? What are the connections of these characteristics to their learning of target idiom knowledge?

5. How do the participants perceive the English idiom learning tasks, the use of text- based online chat for collaboration, as well as the effectiveness of IREs for English idiom learning?

2.6. Chapter Summary

Chapter 2 provides an overview of the few areas that are closely related to the examination of SCMC-based collaborative dialogue and its effects on L2 development.

Specifically the review of literature is organized around the key construct of this study, namely collaborative dialogue or LREs. It gives a detailed account of the definition of collaborative dialogue, its relationship to SLA, factors that impact its emergence, its connection to L2 lexical acquisition, along with its operationalization in SCMC. It also introduces corpus-based analysis of English idioms and the teaching and learning of English idioms. Terminologies central to sociocultural SLA such as mediation,

internalization, scaffolding, and ZPD are accounted for and the way they were deployed for the analysis and interpretation of the data in this study is also elaborated. Issues in current studies relating to collaborative dialogue, L2 lexical acquisition, and SCMC are discussed so that how the current study may fill the gap in the existing literature is clearly revealed. The reasons for the selection of English idioms that are frequently used in academic discourse and the choice of the target idioms are also described. Finally, the research questions governing the current study and their rationale are provided. Chapter 3 introduces the approach to research, the participants, data collection materials and

CHAPTER 3. METHODOLOGY

Chapter 3 provides an overview of the methodology employed in this study, that is, a qualitative, descriptive case study within the theoretical framework of sociocultural SLA. It begins with a thorough explanation of the rationale for the choice of a descriptive case study and proceeds to a detailed description of the research setting, participants, and the role of the researcher. Case selection and sampling are then elaborated. After that, it offers a thorough account of the pedagogical materials, notably the English idiom learning tasks and its rationale. Information is also provided about data collection materials and instruments, including the pre-task questionnaire, pre and posttests,

reflective journals, stimulated recall protocols, as well as post-task survey and interviews. The chapter concludes with an in-depth discussion of the procedures for data collection and the quantitative and qualitative analyses conducted for addressing each research question.

3.1. Approach to Research

This study used a descriptive case study to examine how collaborative dialogue, or more specifically idiom-focused-dialogue (IFD), was carried out in a text-based SCMC environment. Specifically it focused on the patterns of dyadic interaction, the utilization of communication strategies, the provision of scaffolded assistance, the characteristics of collaborative interaction, and participants’ perspectives on their online exchanges. As Yin (2014) described it, “A case study is an empirical inquiry that

investigates a contemporary phenomenon (the “case”) in depth and within its real-world context, especially when the boundaries between phenomenon and context may not be

clearly evident” (p. 16). The choice of a descriptive case study as the main approach to investigation to a large extent was motivated by this statement.

A perusal of the extant literature suggested that although Swain and her co-

researchers introduced the construct of collaborative dialogue to the field of SLA, most of her own and succeeding studies were experimental, quantitative in nature, without

sufficient attention to the contextual factors influencing the production of collaborative dialogue. However, as Johnson (1992) aptly pointed out, “If experimenters focus on inputs and outputs without attending to processes and contexts, they may completely miss the important events that most shaped the results. Without rich descriptions of what goes on, there is little basis for adequate interpretation” (p. 187). Watanabe and Swain (2007) similarly maintained that many of the questions concerning collaborative dialogue

“cannot be answered from the quantitative analysis of LREs alone” (p. 124). Additionally, recent studies on collaborative dialogue revealed that the conduct of LREs, in both face- to-face (e.g., Watanabe, 2008; Dobao, 2012) and real-time (e.g., Lee, 2008; Peterson, 2012) communication, was largely influenced by the contexts of peer-peer collaborative interaction. Due to the fact that “Sociocultural approaches prioritize qualitative research methodology and pay close attention to the settings and participants in interactions” (Foster & Ohta, 2005, p. 403), it appears that the adoption of the type of methodology that is advantageous in examining collaborative dialogue in relation to its emerging context and various sociocultural factors is not only essential but also fundamental to a more complete and comprehensive understanding of languaging in SCMC.

A descriptive case study serves the above-mentioned purpose quite well since it “aims to present a detailed, contextualized picture of a particular phenomenon”

(Heighham & Croker, 2009, p. 71). Unlike the bulk of prior studies focusing on the linkage between episodes of collaborative dialogue and posttest scores, the current study did not seek to verify a causal relationship between collaborative dialogue episodes and L2 vocabulary learning or generalize the results, but rather attempted to give a thorough and clear account of how SLA stemmed from peer-peer interaction and how the co- constructed knowledge about the target English idioms was appropriated and internalized at an individual level within the context of ESL teaching and learning. The detailed description and in-depth analyses required for achieving this research objective dictated the dominant use of a descriptive case study, which intended to “understand the

complexity and dynamic nature of the particular entity, and to discover systematic

connections among experiences, behaviors, and relevant features of the context” (Johnson, 1992, p. 84).

The foregrounding of the case study approach, however, did not preclude the incorporation of a quantitative component. As Tocalli-Beller (2005), justifying her inclusion of quantitative data in her case study, reasoned, “to know if one task, one teaching method, one type of learning environment or one program is more effective than another, quantitative accounts can enrich the observations and qualitative analyses” (p. 59). Other researchers (e.g., Duff, 2008; Eisenhardt, 2002; Yin, 2003) made similar claims about the legitimacy of analyzing cases on the basis of quantitative evidence. Quantitative analyses in the current study were therefore conducted through a pre/posttest design measuring the growth of the participants’ target English idiom knowledge through their collaborative dialogue. Overall, quantitative evidence in this study undertook a “supportive, secondary role” (Creswell & Plano Clark, 2007, p. 67) and complemented

qualitative analysis so that “valid and well-substantiated conclusions” (ibid, p. 65) concerning SCMC-based collaborative dialogue could be drawn.

As the literature suggested, some prominent features of a case study include its boundedness, multiple sources of evidence, and interpretative stance. The descriptive case study design involved in the current study exhibited these features. For boundedness, Merriam (2002) noted that a case is a “bounded, integrated system” (p. 8). Chapelle and Duff (2003) further made the point that “In TESOL, a case typically refers to a person, either a learner or a teacher, or an entity, such as a school, a university, a classroom, or a program” (p. 164). The current study can be viewed as a single-case study in which the class was the case. It was concerned with the development of target idiom knowledge within a college-level academic writing class through a nine-week longitudinal study. Furthermore, as Yin (2009) pointed out, “within a single case, attention is also given to a subunit or subunits” (p. 50), and the subunits are often associated with the primary research questions of the case study (Yin, 2014). The subunits of analysis in this study consisted of the eight dyads undergoing the dialogic process intended to help them attain a full understanding of the meaning of the target idioms and their reactions to this whole learning process.

Additionally, the multiple sources of evidence in the current study were crystalized in the exploitation of a variety of quantitative and qualitative measures. Specifically, quotations and excerpts from interviews, stimulated recalls, chat transcripts, and reflective journals constituted the qualitative data, while descriptive quantitative analyses such as frequency counts, gains scores, and percentages were simultaneously conducted to supplement the qualitative analyses. Quantitative and qualitative data were

also triangulated to aid in the interpretation of the results and increase the validity of the findings.

Finally, for the accuracy of interpretation, Chapelle and Duff (2003) proposed several strategies, including having additional coders, consultation with case participants for their interpretations of data or findings, and the use of L1s for low L2 proficiency participants. This study adhered to these guidelines by incorporating additional coders for quantitative and qualitative data analyses, conducting member checks and peer

debriefings, as well as employing the participants’ L1 (Mandarin Chinese) for interviews and stimulated recalls. These measures, to a large degree, helped establish the

trustworthiness of the case study by making “clear, credible, and convincing” arguments (Chapelle & Duff, 2003, p. 167).

3.2. Research Setting

The current study was conducted in a college-level advanced ESL academic writing course, English 101C: English for Native Speakers of Other Languages (hereafter referred to as ENGL101C), at Iowa State University during the spring semester of 2015. Prior to the data collection for this study, human subject approval was obtained from the university’s Institutional Review Board (IRB) committee (see Appendix J). ENGL101C was a 16-week semester course, and the students and instructor met twice every week: once on Tuesdays and the other time on Thursdays. Classes on Tuesdays were lab sessions in which the students performed writing-related activities and tasks on the computer. Classes on Thursdays were in a face-to-face classroom where the instructor gave lectures and offered writing workshops. On top of classroom meetings, the instructor also held individual conferences with the students for the discussion of the

revisions of the drafts of their writing assignments. Throughout the study, I was the researcher and also the instructor for ENGL101C. There were 16 students enrolled in this class during the period of the current study.

The fitness of ENGL101C to the current study can be accounted for in three ways. First of all, to adequately fulfill the requirements of ENGL101C, the students needed to collaborate on a series of pair work and team projects. Therefore, their ability to engage in intercultural communication and contribute to classroom discussions was essential to their academic success. This study attempted to reveal the details of peer-peer

collaboration when ESL learners were faced with cognitive demanding academic tasks. The difficulties and challenges that the participants experienced during collaborative interaction were thus useful for enhancing their collaborative learning. In addition, to successfully complete the writing assignments, it was important for the students to fully understand the content of the textbook and reading materials. Due to the frequent use of English idioms in academic discourse, knowledge about the meaning of the target idioms would help student comprehend the text they encountered in their study. This was

particularly true with one ENGL101C assignment in which the students were required to watch an English movie together and create a website about it. Their grasp of the

meaning of English idioms could undoubtedly help them understand the storyline of the movie and write a movie review of a better quality together. Furthermore, proper application of English idioms in the students’ essays could make their writing more native-like and their communication more effective. Finally, on some occasions, students in ENGL101C were asked to use online chat for interaction. For example, since it was impractical for the students to watch the movie together during class time, they were

instructed to share their thoughts and reactions through online chat. The examination of their real-time discussions that the current study was targeted at was therefore essential for a more comprehensive understanding of the students’ use of this tool for their L2 learning.

3.3. Participants

All 16 students in ENGL101C that I taught agreed to participate in this study. These students were placed in the class according to their scores on the writing section of the English placement test they took upon entering the university. Through contact with the participants for the first few weeks of the semester and their responses to the pre-task questionnaire, I had gained some preliminary knowledge of their demographic

information. The participants’ English proficiency was high intermediate, as determined by their self-reported TOEFL or IELTS scores (70 or above in TOEFL iBT or 6 or above in IELTS). Their average age was 19.9 years (range 18-22, SD = 1.25), with an average duration of residence in the US of 7.6 months (range 1-14, SD = 6.04) and an average length of English learning of 11.6 years (range 7-18, SD = 5.32). Since most of them had been in the United States for only a short time, they were adjusting to the new academic and social demands of college. Among the sixteen participants, nine were from the college of business (business, finance, and management), five were engineering majors (civil and computer), one was in nutritional science, and one was in chemistry. Eight of the participants were Chinese, one Malaysian, two Koreans, one Turk, one Bangladeshi and three Emiratis. Detailed demographic information about the participants is listed in Table 2.

Table 2. Demographic information of the participants (N=16)

*Note. Throughout the study, the stimulated recalls and post-task interviews were conducted in Mandarin Chinese for all Mandarin-speaking participants. For participants of other L1s, the stimulated recalls and post-task interviews were conducted in English.

Participants L1 Gender Age English

Proficiency Test taken Test score Major Months in US Years of English Instruction A B C D E F G H I J K L M N O P Mandarin Mandarin Mandarin Mandarin Mandarin Cantonese Mandarin Arabic Bangla Turkish Mandarin Mandarin Malay Arabic Korean Arabic Male Male Female Female Female Female Male Male Male Female Female Female Female Male Male Male 21 22 18 20 20 18 20 19 19 20 21 20 21 18 21 20 iBT iBT iBT IELTS IELTS IELTS IELTS IELTS SAT critical reading IELTS IELTS IELTS IELTS PBT iBT IELTS 95 98 72 6 6 7.5 6 6 510 6 6 6 7 537 76 6 Mechanical Engineering Chemistry Business Business Business Nutritional Science Finance Accounting Computer Engineering Business Electrical Engineering Accounting Software Engineering Management Management Finance 36 18 6 3 3 6 8 18 2 7 2 19 36 8 24 16 5 5 12 7 5 14 9 12 11 14 12 12 21 9 8 10

her doctoral dissertation on ESL dyads’ collaborative dialogue concerning English puns, made the point that it was the adequacy of the sample rather than the number of

participants that mattered most to qualitative research. In particular, she, echoing Morse (1998), maintained that adequacy involved the collection of “a sufficient amount of data…so that the phenomena begin to repeat themselves” (p. 44) and explained the adequacy of her data in terms of the number and characteristics of the puns in the tasks she employed. To some extent, her concept of adequacy helped justify the number of participants in this dissertation. In total, the eight dyads collaborating on eight idioms across the four English idiom-learning tasks generated 256 IFD episodes, which appeared to be sufficient for the emergence of patterns of interaction, communication strategies, collective scaffolding, and microgenetic episodes. Likewise, each idiom elicited 16 instances of IFD so that how languaging about lexical issues in academic contexts took place can be adequately examined. Additionally, stimulated recalls, along with post-task interviews, generated approximately 10 hours of audio recordings and 30 pages of transcriptions, which were enough to draw sound conclusions about the participants’ perceptions of and attitudes toward SCMC-based collaborative dialogue.